Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


OCR-D wrapper for ocr-fileformat



  • GNU make
  • Python && pip
  • OpenJDK (required by submodule)
  • optional: Docker CE for building container images


Clone the repository and it's submodule recursive:

git clone --recursive

Step into local clone, build and install ocr-fileformat and the ocrd_fileformat OCR-D wrapper:

make -C ocrd_fileformat install

Alternatively, for the Docker option, just get:

docker pull ocrd/fileformat


After successful installation type ocrd-fileformat-transform --help to get an idea which conversions are supported already:

ocrd-fileformat-transform -h
Usage: ocrd-fileformat-transform [OPTIONS]

Convert between OCR file formats

> Processor base class and helper functions. A processor is a tool > that implements the uniform OCR-D command-line interface for run- > time data processing. That is, it executes a single workflow step, > or a combination of workflow steps, on the workspace (represented by > local METS). It reads input files for all or requested physical > pages of the input fileGrp(s), and writes output files for them into > the output fileGrp(s). It may take a number of optional or > mandatory parameters. Process the :py:attr:workspace from the > given :py:attr:input_file_grp to the given > :py:attr:output_file_grp for the given :py:attr:page_id under > the given :py:attr:parameter.

> (This contains the main functionality and needs to be overridden by > subclasses.)

Options: -I, --input-file-grp USE File group(s) used as input -O, --output-file-grp USE File group(s) used as output -g, --page-id ID Physical page ID(s) to process --overwrite Remove existing output pages/images (with --page-id, remove only those) -p, --parameter JSON-PATH Parameters, either verbatim JSON string or JSON file path -P, --param-override KEY VAL Override a single JSON object key-value pair, taking precedence over --parameter -s, --server HOST PORT WORKERS Run web server instead of one-shot processing (shifts mets/working-dir/page-id options to HTTP request arguments); pass network interface to bind to, TCP port, number of worker processes -m, --mets URL-PATH URL or file path of METS to process -w, --working-dir PATH Working directory of local workspace -l, --log-level [OFF|ERROR|WARN|INFO|DEBUG|TRACE] Log level -C, --show-resource RESNAME Dump the content of processor resource RESNAME -L, --list-resources List names of processor resources -J, --dump-json Dump tool description as JSON and exit -h, --help This help message -V, --version Show version

Parameters: "from-to" [string - "page alto"] Transformation scenario, see ocr-fileformat -L Possible values: ["abbyy hocr", "abbyy page", "alto2.0 alto3.0", "alto2.0 alto3.1", "alto2.0 hocr", "alto2.1 alto3.0", "alto2.1 alto3.1", "alto2.1 hocr", "alto page", "alto text", "gcv hocr", "gcv page", "hocr alto2.0", "hocr alto2.1", "hocr page", "hocr text", "page alto", "page hocr", "page page2019", "page text", "tei hocr"] "ext" [string - ""] Output extension. Set to empty string to derive extension from the media type. "script-args" [string - ""] Arguments to Saxon (for XSLT transformations) or to transformation script

With the OCR-D CLI wrapper the ocr-fileformat converter integrates fluently into existing OCR-D tool workflows.

Given a previous step which produces PAGE-XML under the file group OCR, a conversion into plain text under the file group OCR-TXT can be achieved with:

ocrd-fileformat-transform -I OCR -O OCR-TXT -P from-to "page text"

With bertsky/workflow-configuration

OCR-TXT: TOOL = ocrd-fileformat-transform
OCR-TXT: PARAMS = "from-to": "page text"

Since the conversion from PAGE-XML to ALTO-XML (V4.1) is such a common requirement, it is the default value for the parameter from-to. Therefore, parameters can be omitted completely:

ocrd-fileformat-transform -I OCR -O OCR-ALTO

With bertsky/workflow-configuration

OCR-ALTO: TOOL = ocrd-fileformat-transform

However, typically the ALTO converter itself will require additional parameters to be able to cope with the kind of annotations present. For example, if you have no cropping in the workflow, and OCR text is only annotated on the line level, then you will need to add:

ocrd-fileformat-transform -I OCR -O OCR-ALTO -P script-args "--no-check-border --no-check-words --dummy-word"

With bertsky/workflow-configuration

OCR-ALTO: TOOL = ocrd-fileformat-transform
OCR-ALTO: PARAMS = "script-args": "--no-check-border --no-check-words --dummy-word"

To run the program via Docker, just spin up a container analogously:

docker run --rm -v $PWD:/data ocrd/fileformat ocrd-fileformat-transform -I OCR -O OCR-ALTO